• This project demonstrates a real-time hand gesture recognition system using Python, OpenCV, and Gemini AI by Google.
• The system allows users to solve math problems by drawing them in the air.
• The repository contains all the code and resources needed to implement this project.

• Real-time hand detection and tracking using OpenCV and cvzone.
• Recognition of specific hand gestures to perform actions like drawing, clearing canvas, and requesting math problem solutions.
• Integration with Gemini AI for solving math problems based on gestures.

• Python 3.10.12
• cvzone pip install cvzone
• OpenCV pip install opencv-python
• Gemini AI pip install google-generativeai
• Python packages
  • numpy pip install numpy
  • pillow pip install pillow

• C++ compiler - install Visual Studio

1. Clone the repository:

   git clone https://github.com/LasithaAmarasinghe/Hand-Gesture-Math-Solver.git
   cd Hand-Gesture-Math-Solver

2. Create and activate a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate   # On Windows use `venv\Scripts\activate`

3. Install the required packages:

   pip install -r requirements.txt

1. Get a Gemini API key

   You can get an API key from here

2. Replace API key in the code with the obtained API key

3. Execute the main script

python math_solver.py

4. Interact with the Application

   • Live Feed
     • The application will display the live feed from your webcam.
   • Perform Hand Gestures
     • Draw: Use one finger to draw on the canvas.
     • Clear: Use one finger to clear the canvas.
     • Solve Math Problem: Use all fingers except the thumb to submit a math problem for solving.

1. getHandInfo(img)

   • Purpose: Detects and tracks the user's hand in each frame of the webcam feed.
   • Implementation: Uses the HandTrackingModule to identify and monitor hand movements.

2. draw(info, prev_pos, canvas)

   • Purpose: Draws on the canvas based on the detected hand gestures.
   • Parameters:
     • info: Information about the detected hand gestures.
     • prev_pos: Previous position of the drawing.
     • canvas: Canvas to draw on.

3. sendToAI(model, canvas, fingers)

   • Purpose: Sends the drawn content to Google's Generative AI for math problem solving based on recognized gestures.
   • Parameters:
     • model: Instance of the Generative AI model.
     • canvas: Canvas containing the drawn content.
     • fingers: Array indicating the state of fingers (e.g., for gesture recognition).

Left-Right Hand Detection

Hand Gesture Detection

Math Expression Detection & Canvas

• This project is licensed under the MIT License. See the LICENSE file for details.